I have a model of depression which I am attempting to convert into a functional simulation in the form of a recurrent neural network or genetic algorithm, whichever would suit it better. (The model is attached)

Some of my questions on how this may work include:

  • How ought one optimise the initial parameters of the system?
  • How best to deal with positive feedback-loop induced instability?
  • How to go about choosing between activation functions
  • Any suggestions on how to go about coding an initial loop and then expanding it would be very helpful. Also, any suggestions on where to code this, I was thinking Jupyter.

    More Chris Kontis's questions See All
    Similar questions and discussions